On Thu, Oct 31, 2002 at 11:51:09AM -0500, Joe Rioux wrote:
> I have two followup questions relative to this.
> 1) Someone mentioned that I really shouldn't simply
> run the copy command through a system call, but
> instead use something akin to system("cmd.exe copy ...")
> since copy is internal to it. I asked him why but
> didn't get a response. I assumed, perhaps incorrectly,
> that a system() call inherits the shell of the parent
> process and would therefore simply run the internal
> copy command for cmd.exe.
> Does anyone know why he said I shouldn't do this?
system(SCALAR_ARG) ends up running "cmd.exe /x/c SCALAR_ARG" on
>=WINNT and "command.com /c SCALAR_ARG" on <WINNT. Specifying cmd.exe
would be like calling "cmd.exe /x/c cmd.exe ...". This is the same on
Unix, in that system(SCALAR_ARG) on UNIx runs "/bin/sh -c
'SCALAR_ARG'". The same thing applies - specifying /bin/sh explicitly
would be like calling "/bin/sh -c '/bin/sh -c \'SCALAR_ARG\''". It
will work, but is is unnecessary.
The person who suggested that you need to specify cmd.exe
may not have been aware of this.
> 2) Perhaps it would be better for me to issue the
> copy command then rather than check the exit
> code use the -f operator to see if the resulting
> file exists. Does that seem reasonable?
It would be better to use File::Copy and not rely on external non-portable
commands.
mark
--
[EMAIL PROTECTED][EMAIL PROTECTED][EMAIL PROTECTED] __________________________
. . _ ._ . . .__ . . ._. .__ . . . .__ | Neighbourhood Coder
|\/| |_| |_| |/ |_ |\/| | |_ | |/ |_ |
| | | | | \ | \ |__ . | | .|. |__ |__ | \ |__ | Ottawa, Ontario, Canada
One ring to rule them all, one ring to find them, one ring to bring them all
and in the darkness bind them...
http://mark.mielke.cc/
_______________________________________________
ActivePerl mailing list
[EMAIL PROTECTED]
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
